PII-032 - MOVING TOWARD "FIT-FOR-PURPOSE" QUANTITATIVE SYSTEMS PHARMACOLOGY (QSP) MODELING IN DRUG DEVELOPMENT: CASE EXAMPLES IN ONCOLOGY.

K. Okuneye¹, Y. Li¹, J. Sheng¹; ¹Incyte Research Institute, Wilmington, DE, USA.

Background: QSP models are emerging as pivotal tools in Model-Informed Drug Development for oncology, aiding dose selection and optimization, and understanding the tumor microenvironment. We propose a framework for "fit-for-purpose" QSP models, emphasizing practical applicability for the pharmaceutical and biotech sectors. We anticipate that the fit-for-purpose QSP models would evolve, thus enhancing their practical value in drug development. This study integrates insights from QSP models across academia, industry, and regulatory agencies to explore how these models can balance complexity with practicality.

Methods: We conducted a 10-year (2014-2024) literature review of QSP models in oncology from PubMed, Scopus, Conference proceedings, regulatory guidance, and publications. The QSP models consists of mechanistic pharmacokinetic pharmacodynamic models to large-scale disease platform models. Representative case studies were selected based on model scope, calibration data, validation processes, and the key questions they addressed. We analyzed these cases to evaluate the alignment between model complexity and their specific objectives.

Results: More than 30 published models from multiple databases, including PubMed and Scopus, were evaluated for suitability, and a subset of these studies was included for final analysis. Despite the diversity of QSP models reviewed, many shared foundational sub-models, with academia laying the groundwork for industry. For example, a QSP model platform was developed to model the complex interactions between immune cells, cancer cells, immunotherapy, and other factors (Sové RJ, et al. CPT Pharmacometrics Syst Pharmacol. 2020). This platform is used to assess different immunotherapy combinations and predict patient outcomes through virtual trials. However, complex models like that of Sové et al. can be effectively reduced to a “fit-for-purpose” model while still preserving the essential biological mechanism and predictive capabilities, especially in industry where there are limited data to validate the QSP model as described in Lemaire V, et al (Clin Pharmacol Ther. 2022).

Conclusion: The “fit-for-purpose” QSP models would align tightly with industry needs, regulatory expectations, and the constraints of drug development timelines and resources.